Title :
Neural networks based current regulator for PWM inverters
Author :
Kazmierkowski, M.P. ; Viet, P.H. ; Dzieniakowski, M.A.
Author_Institution :
Inst. of Control & Ind. Electron., Warsaw Univ. of Technol., Poland
Abstract :
This paper presents a neural network (NN) based current regulator for transistor PWM inverters. The neural network operates as a digital filter and tunes the controller gain. This NN regulator has the potential of replacing the linear PI regulator which is traditionally applied in the system with PWM modulators. The results of comparative study which illustrate the performance of the NN regulators in comparison to the systems with a PI regulator are presented. Also, learning possibilities and robustness to parameter change are discussed
Keywords :
control system synthesis; digital filters; electric current control; invertors; neural nets; pulse width modulation; PWM inverters; current regulator; digital filter; electric current control; gain; neural network; performance; power transistors; robustness;
Conference_Titel :
Power Electronics and Applications, 1993., Fifth European Conference on
Conference_Location :
Brighton
